The invention relates to a compensation circuit for the correction of image defects of a television image, in particular an eidophor image projection system, wherein an electron beam controls the luminous flux of a strong source of light via a schlieren optics. In such a compensation circuit, two line memories, at least one coefficient multiplier, a first adder to generate a compensation signal of three consecutive lines and a second adder for the addition of the compensation signal to the image signal of the intermediate line of three adjacent lines, are provided for the video signal of every displayable color extract or black/white image.
A compensation circuit of this type is described in EP-A1-258 883 and is used to suppress defects in the image projected. The defects are caused by a storage or edge pile-up when, between two adjacent lines of the projection image, a transition from dark lines or line segments to bright lines or bright line segments is taking place. Particularly in a high definition television system, the excess brightness occurring at line packet edges due to the deformation of the control layer in the form of an oil film is disturbing. To eliminate disturbances at horizontal edges, vertical filtering is effected in the known compensation circuit, whereby the stepped edge of the input signal of the vertical filter is converted into a gradual transition. This, however, leads not only to the suppression of the occurrence of bright disturbance effects at horizontal edges by filtering in the vertical direction, but also to a smearing of the horizontal edge in the vertical direction because of the variation of space frequencies. The resulting loss of resolution is especially disturbing in a high definition television system.
In the known compensation circuit the first adder has three inlets for the feeding in of the video signals of three adjacent lines. The video signals are fed in via coefficient multipliers, which multiply the highest and the lowest of the three signals by a constant factor of 0.25. The video signal of the intermediate line is multiplied by a constant factor and an associated coefficient multiplier. The coefficient multiplier causes the constant factor of the intermediate line to correspond to twice the negative value of the compensation factor used for the lowest and highest lines.
It is therefore an object of the present invention to provide a compensation circuit with improved vertical resolution so that finer periodic vertical structures may also be displayed as a gray area, without the defect of overshooting or differentiation, or smearing.